Although it is well-established that alcohol and tobacco use are linked, relatively little is known about the nature of the relationship, as most research has focused on one substance or the other. The purpose of the present study is to explore potential differences at different stages of SATU in terms of risk factors and reactivity t environmental substance cues and to develop a measure of motives and expectancies for simultaneous use of alcohol and tobacco (SATU). Utilizing items drawn from existing alcohol- and tobacco-specific measures and [previously tested in focus group discussions], we will develop a measure of motives and expectancies for alcohol SATU. Next, we will recruit 18-25 year olds (n = 300) from the community. Subjects will be divided evenly into four groups based on current alcohol and tobacco use: heavy/heavy, heavy/light, light/heavy, and light/light. Subjects will be enrolled and complete the new measure as well as assessment of a number of risk factors online. Finally, 60 subjects (15 from each group, matched on age, sex, race/ethnicity) will be selected to participate in an fMRI session in which they will be exposed to alcohol, tobacco, [and SATU] cues. Cue reactivity will be measured in terms of brain activity and self-reported craving for alcohol and tobacco. Psychometric properties of the new measure of co-use motives and expectancies will be assessed via exploratory [and confirmatory] factor analysis. We anticipate several distinct but related factors, including enhancement, automaticity, and social/conformity motives. We will also evaluate the validity of the new scale by comparing scores among the groups, and correlating scores with the assessed risk factors. We hypothesize that the two heavy alcohol use groups will exhibit the most positive expectancies and strongest motives for co-use, followed by the two light alcohol groups. It is also expected that scores will be significantly correlated with frequency of SATU and with several risk factors for substance use. Finally, we predict that both heavy alcohol groups will report more positive expectancies for SATU than for either substance alone, and that their expectancies for SATU will be more positive than the other groups. In terms of the fMRI task, we expect that heavy alcohol users will exhibit greater BOLD and craving reactivity to alcohol-only cues compared with others. We also hypothesize that heavier co-users will exhibit cross-reactivity, such that alcohol-only cues elicit tobacco as well as alcohol craving, and vice versa. Finally, we expect that SATU cues will elicit greater BOLD and craving responses than alcohol-only cues in those who report frequent simultaneous use.